Table 4.

Binary Logistic Regression Results Predicting Diagnostic Status (MDD, HC) From ERPs

Model With Predictors	Diagnostic Status (MDD, HC)
	R2	χ2	OR (95% CI)	p Value	VIF
Model 1
Effort-P3	0.65	42.9a	–	–	–
High Effortb	–	–	0.85 (0.72–0.99)	.038	1.59
Low Effort	–	–	0.87 (0.73–1.04)	.137	1.60
Age	–	–	0.96 (0.91–1.02)	.168	1.02
Model 2
SPN	0.22	11.8c	–	–	–
High Effort	–	–	0.81 (0.62–1.05)	.114	2.41
Low Effortb	–	–	1.31 (1.0003–1.71)	.0497	2.42
Agec	–	–	0.94 (0.90–0.98)	.007	1.05
Model 3
RewP	0.26	13.5c	–	–	–
High Effortb	–	–	0.89 (0.80–0.99)	.039	1.30
Low Effort	–	–	1.03 (0.94–1.13)	.513	1.09
Agec	–	–	0.93 (0.88–0.97)	.002	1.21
Model 4
Feedback-P3	0.24	12.5c	–	–	–
High Effort	–	–	1.03 (0.88–1.20)	.755	7.05
Low Effort	–	–	0.92 (0.79–1.09)	.330	7.08
Agec	–	–	0.94 (0.90–0.99)	.009	1.01

Logistic regression was used to predict depression diagnostic status (0 = HC, 1 = MDD).

The Nagelkerke R² and χ² statistics are reported for the logistic regression model and reflect statistics comparing the full model to the null model.

HC, healthy control; MDD, major depressive disorder; OR, odds ratio; RewP, reward positivity; SPN, stimulus-preceding negativity; VIF, variance inflation factor.

^ap < .001.

^bp < .05.

^cp < .01.